Unraveling the roles of excessive alcohol use and liver disease in mortality

Excessive consumption of alcoholic beverages has long been recognized as the major cause of death from liver disease. During the height of the COVID-19 pandemic, the health consequences of excessive drinking became even more obvious as demonstrated by a 25% increase in alcohol-related deaths and a 22% increase in alcohol-associated liver disease (ALD) in the United States from 2019 to 2020 (White et al., 2022). The article by Whitfield, Seth, Morgan and others provides confirmation of the health-related consequences of excessive drinking in a large population-based cohort, the U.K. Biobank (Whitfield et al., 2022). Their study clearly indicates an increase in both all-cause and liver-related mortality in excessive drinkers defined as men drinking more than 80 g and women drinking more than 50 g of ethanol daily for at least 10 years. Beyond confirming, with substantial statistical power, the harm associated with excessive consumption, their study calls attention to issues that deserve further research and clinical situations that provide an opportunity for intervention.

In discussing the findings from this important study, we must point out that the focus was on risk factors for all causes of mortality only in excessive drinkers compared to the rest of the population, not the impact of all ranges of alcohol consumption on all-cause mortality. While important, the relationship between alcohol consumption and all-cause mortality has been studied extensively with conflicting results regarding the effects of moderate drinking on mortality (Collaborators, 2018, Dawson, 2000, Di Castelnuovo et al., 2006, Xi et al., 2017). Most previous population studies found an increase in all-cause mortality in excessive drinkers. However, many of those studies lacked enough excessive drinkers within the study population to determine factors that influence the mortality from liver diseases, cancer, heart disease and other health problems.

Whitfield and colleagues found that all-cause mortality in both men and women who drank excessively was double that of the rest of the population (Whitfield et al., 2022). Not surprisingly, the mortality from liver disease in excessive drinkers was significantly higher than for the rest of the population with a hazard ratio 6.67 after adjustment for smoking and other risk factors including diabetes, BMI, waist-hip ratio and socioeconomic factors. The adjusted mortality hazard ratio for cardiovascular disease (CVD), malignancies, respiratory and infectious diseases was also significantly higher in excessive drinkers. However, with the exception of a higher percentage of deaths being due to liver disease, the percentage of deaths from all other causes such as heart disease, cancer, respiratory diseases and infections was similar for excessive drinkers and the rest of the population (see Figure 2, (Whitfield et al., 2022)). One important limitation of the study, as noted by the authors, is that excessive drinkers who died before age 40 were not included in the U.K. Biobank database so premature deaths due to trauma and ALD in younger individuals were not captured. During the height of the COVID pandemic, deaths from liver disease and other alcohol-associated diseases increased in individuals under 40 years of age (White et al., 2022).

All-cause mortality has been shown to be increased in patients with ALD (adjusted hazard ratio of 4.70) (Hagstrom et al., 2021). In the U.K. Biobank study, Whitfield and colleagues found elevations in AST, GGT and alkaline phosphatase were associated with an increase in both all-cause and liver-related mortality in excessive drinkers (Whitfield et al., 2022). These important findings suggest that all excessive drinkers should be screened carefully for evidence of liver disease. Liver disease is usually asymptomatic until the later stages of the disease process which means that biochemical tests and/or imaging are required to identify patients with early-stage disease that is more amenable to treatment than advanced cirrhosis. Elevations in liver enzymes can be used to detect the presence of liver disease although the degree of elevation of AST, GGT and alkaline phosphatase do not correlate well with severity of fibrosis in patients with liver disease. The observation that elevations in these enzymes are biomarkers for all-cause mortality deserves further study.

Increases in liver enzymes were also associated with higher risk of cancer-related and cardiovascular mortality (see Supplementary Table 7, (Whitfield et al., 2022)). In addition, ALD was a risk factor both for cancer and for CVD in excessive drinkers in univariate, but not multivariate analysis (Whitfield et al., 2022). At first glance, these associations may seem surprising, but a recently published Swedish population based cohort study (ESPRESSO) found that patients with biopsy-proven ALD experienced earlier development of cardiovascular events, particularly in the first year after diagnosis, compared to the rest of the population even after adjusting for diabetes and obesity (Hagstrom et al., 2022b). Other studies identified a similar increase in risk of CVD in patients with ALD (Chang et al., 2019). The risk of cancer, particularly hepatocellular carcinoma (HCC) was increased in patients with biopsy-proven ALD, although death from causes other than cancer diminished the significance of the association (Hagstrom et al., 2022a).

Increased mortality from CVD and cancer has also been noted in patients with non-alcoholic fatty liver disease (NAFLD). A meta-analysis of 10 studies with 182,000 participants found a 1.5–2.0-fold increase in the hazard ratio for extrahepatic gastrointestinal cancers (esophagus, stomach, pancreas, colon) and a 1.2–1.5-fold increase for lung, breast and genitourinary cancers in patients with NAFLD independent of age, sex, smoking, obesity, diabetes and other potential confounders (Mantovani et al., 2022). Unfortunately, the studies included in this meta-analysis did not all report the severity of NAFLD by stage of fibrosis, so linking extrahepatic cancers to severity of liver disease was not possible. Another large meta-analysis of 36 studies with almost 6 million participants found a strong association between NAFLD and risk of both fatal and nonfatal cardiovascular disease (CVD) (Mantovani et al., 2021). In this report, the stage of NAFLD fibrosis was a strong predictor of the risk of CVD events (hazard ratio 2.50) independent of age, sex, adiposity, presence of diabetes, serum lipids and other known cardiovascular predictors.

The question of whether low to moderate amounts of alcohol contribute to progression of liver disease in patients with NAFLD has not been settled. In a meta-analysis, Rehm and colleagues showed that the risk of liver disease is increased in both men and women at levels of alcohol consumption that are less than the upper allowance of drinks/day in most medical society definitions of NAFLD (Rehm et al., 2010). Since genetic risk factors are similar for both NAFLD and alcohol-associated fatty liver disease, the increased risk of liver disease with moderate daily consumption may be related to alcohol combined with risk factors for NAFLD such as obesity and diabetes (Schwantes-An et al., 2021, Kubiliun et al., 2022).

The reported increase in risk for cancer and CVD in the U.K. Biobank study combined with an increased risk of CVD and cancer in patients with NAFLD raise intriguing questions about the fundamental role of liver disease as a risk factor for other diseases. Further study will certainly be necessary to determine whether this association is real or related to confounding from alcohol consumption, socioeconomic, or other factors that are difficult to measure. Clarifying the contribution of liver disease to other chronic diseases is essential.

The U.K. Biobank study clearly shows the strong relationship between quantity of alcohol consumed and the increased risk of all-cause in excessive drinkers, both men and women (Whitfield et al., 2022). One of the advantages of the U.K. Biobank study is that the excessive drinkers were selected from the general population, not because they were seeking treatment. In this study, not only the quantity of alcohol consumed but also an ICD-10 diagnosis of alcohol dependence (F10.2) or harmful use of alcohol were strong risk factors for all-cause mortality in excessive drinkers. Alcohol dependence was diagnosed in 6.7% of excessive drinkers in the U.K. Biobank study. The association of alcohol dependence with all-cause mortality has been observed in other studies as well (Whitfield et al., 2018, Dawson, 2000). To have a diagnosis of alcohol dependence or other ICD-10 diagnosis of harmful use required the subjects to have been seen in the healthcare system (Whitfield et al., 2022). The reason for the visits may have been for routine or emergency care as well as treatment of an alcohol use disorder (AUD). Some patients may have been recognized as having an AUD during care for a medical problem that was not specifically alcohol-related. Defining the reason for the healthcare visits that led to a diagnosis of alcohol problems might provide interesting insights into how patients with alcohol use disorders present.

Knowing that alcohol consumption is a modifiable risk factor for development of liver disease and that ALD may be a risk factor for overall mortality, CVD and possibly extrahepatic cancers, healthcare providers should be encouraged to assess alcohol consumption not only during treatment for an AUD but also in all other healthcare encounters. Self-administered questionnaires such as AUDIT can detect potentially harmful levels of consumption and timeline-follow-back as well as new technologies such as smartphone apps can quantify drinking (Force et al., 2018). These tools can guide clinicians as they make decisions about screening for diseases as well as potential interventions. A large body of evidence suggests that counseling, even brief intervention is effective in reducing harm associated with ongoing excessive consumption of alcohol (Kaner et al., 2018, Force et al., 2018).

The U.K. Biobank study of excessive drinkers along with other recent studies indicate a strong need to intensify screening for liver disease, CVD and cancer in excessive drinkers. The U.S. Preventive Services Task Force (USPTF) recommends screening for unhealthy alcohol use in primary care populations (Force et al., 2018). However, excessive alcohol consumption is not mentioned as a risk factor in the recent statement from USPTF regarding cardiovascular diseases (Force, 2022) and no mention is made of excessive drinking in the recommendations regarding screening for most common causes of cancer. Current population-based screening strategies for CVD and cancer stratify the population based on family history of cancer or heart disease and individual risk factors such as cholesterol and smoking, but none of the guidelines include alcohol consumption or liver disease as a contribution to overall risk of these diseases.

Some of the findings in the U.K. Biobank study of excessive drinkers, such as the increased mortality from liver disease might have been expected. However, the size of the study, the association between liver disease and other chronic illnesses and the role of alcohol use disorders as independent risk factors not only for liver disease but also for the most common causes of death such as CVD and cancer emphasize the need for clinicians to assess alcohol use and to reduce the potential harm associated with excessive use.